Developing Requirements for Standards in the Field of Ethics of Artificial Intelligence

The research is devoted to the issue of creating a safe and reliable artificial intelligence (AI) system. Objective: to create a clear list of requirements for future standards in the field of ethics of artificial intelligence in order to bridge the gap between theory and practice. Methods: methods of induction and synthesis, comparative analysis and analogy are used. Results: an analysis of the existing guidelines on AI ethics was carried out and it was concluded that it contains mainly general recommendations, without specific technical rules. Possible errors and the reasons for the difficulty of holding potentially responsible parties accountable in the field of AI implementation are presented. A three-component list of requirements for artificial intelligence technologies and the relevant data on the basis of which they are defined has been developed; a step-by-step plan for integrating ethical principles into practical aspects of development and regulation has been proposed. The practical significance lies in offering a framework for national and international regulatory documents that will ensure the safe use of the technology and increase user confidence.